Maria Drangova

In Vivo Small Animal Imaging using Micro-CT and Digital Subtraction Angiography

Small-animal imaging has a critical role in phenotyping, drug discovery and in providing a basic understanding of mechanisms of disease. Translating imaging methods from humans to small animals is not an easy task. The purpose of this work is to review in vivo x-ray based small-animal imaging, with a focus on in vivo micro-computed tomography (micro-CT) and digital subtraction angiography (DSA). We present the principles, technologies, image quality parameters and types of applications. We show that both methods can be used not only to provide morphological, but also functional information, such as cardiac function estimation or perfusion. Compared to other modalities, x-ray based imaging is usually regarded as being able to provide higher throughput at lower cost and adequate resolution. The limitations are usually associated with the relatively poor contrast mechanisms and potential radiation damage due to ionizing radiation, although the use of contrast agents and careful design of studies can address these limitations. We hope that the information will effectively address how x-ray based imaging can be exploited for successful in vivo preclinical imaging.

Prediction of Arrhythmic Events in Ischemic and Dilated Cardiomyopathy Patients Referred for Implantable Cardiac Defibrillator Evaluation of Multiple Scar Quantification Measures for Late Gadolinium Enhancement Magnetic Resonance Imaging

Background— Scar signal quantification using late gadolinium enhancement cardiac magnetic resonance (LGE-CMR) identifies patients at higher risk of future events, both in ischemic cardiomyopathy (ICM) and nonischemic dilated cardiomyopathy (DCM). However, the ability of scar signal burden to predict events in such patient groups at the time of referral for implantable cardioverter-defibrillator (ICD) has not been well explored. This study evaluates the predictive use of multiple scar quantification measures in ICM and DCM patients being referred for ICD. Methods and Results— One hundred twenty-four consecutive patients referred for ICD therapy (59 with ICM and 65 with DCM) underwent a standardized LGE-CMR protocol with blinded, multithreshold scar signal quantification and, for those with ICM, peri-infarct signal quantification. Patients were followed prospectively for the primary combined outcome of appropriate ICD therapy, survived cardiac arrest, or sudden cardiac death. At a mean follow-up of 632 ± 262 days, 18 patients (15%) had suffered the primary outcome. Total scar was significantly higher among those suffering a primary outcome, a relationship maintained within each cardiomyopathy cohort (P<0.01 for all comparisons). Total scar was the strongest independent predictor of the primary outcome and demonstrated a negative predictive value of 86%. In the ICM subcohort, peri-infarct signal showed only a nonsignificant trend toward elevation among those having a primary end point. Conclusions— Myocardial scar quantification by LGE-CMR predicts arrhythmic events in patients being evaluated for ICD eligibility irrespective of cardiomyopathy etiology.

Design and Performance Evaluation of a Remote Catheter Navigation System

A novel remote catheter navigation system has been developed to reduce physical stress and irradiation to the interventionalist during fluoroscopic X-ray guided catheter intervention. The unique teleoperated design of this system allows the interventionalist to apply conventional axial and radial motion, as used in current practice, to an input catheter placed in a radiation-safe location to control a second catheter placed inside the procedure room. A catheter sensor (used to measure motion of the input catheter) and a catheter manipulator (used to manipulate the second catheter) are both presented. Performance evaluation of the system was assessed by first conducting bench-top experiments to quantify accuracy and precision of both sensed and replicated motion, and then conducting two experiments to evaluate the latency from sensed to replicated motion. The first study consisted of replicating motions of prescribed motion trajectories, while the second study utilized eight operators to remotely navigate a catheter through a normal carotid model. The results show the system has the ability to sense and replicate motion to within 1 mm and 1deg in the axial and radial directions, respectively. Remote catheter manipulation was found to be operator dependent and occurred under 300 ms. Future applications of this technology are then presented.

Validation of dynamic heart models obtained using non-linear registration for virtual reality training, planning, and guidance of minimally invasive cardiac surgeries

Current minimally invasive techniques for beating heart surgery are associated with three major limitations: the shortage of realistic and safe training methods, the process of selecting port locations for optimal target coverage from X-rays and angiograms, and the sole use of the endoscope for instrument navigation in a dynamic and confined 3D environment. To supplement the current surgery training, planning and guidance methods, we continue to develop our Virtual Cardiac Surgery Planning environment (VCSP) -- a virtual reality, patient-specific, thoracic cavity model derived from 3D pre-procedural images. In this work, we create and validate dynamic models of the heart and its components. A static model is first generated by segmenting one of the image frames in a given 4D data set. The dynamics of this model are then extracted from the remaining image frames using a non-linear, intensity-based registration algorithm with a choice of six different similarity metrics. The algorithm is validated on an artificial CT image set created using an excised porcine heart, on CT images of canine subjects, and on MR images of human volunteers. We found that with the appropriate choice of similarity metric, our algorithm extracts the motion of the epicardial surface in CT images, or of the myocardium, right atrium, right ventricle, aorta, left atrium, pulmonary arteries, vena cava and epicardial surface in MR images, with a root mean square error in the 1 mm range. These results indicate that our method of modeling the motion of the heart is easily adaptable and sufficiently accurate to meet the requirements for reliable cardiac surgery training, planning, and guidance.

2D-3D registration of coronary angiograms for cardiac procedure planning and guidance

We present a completely automated 2D-3D registration technique that accurately maps a patient-specific heart model, created from preoperative images, to the patients orientation in the operating room. This mapping is based on the registration of preoperatively acquired 3D vascular data with intraoperatively acquired angiograms. Registration using both single and dual-plane angiograms is explored using simulated but realistic datasets that were created from clinical images. Heart deformations and cardiac phase mismatches are taken into account in our validation using a digital 4D human heart model. In an ideal situation where the pre- and intraoperative images were acquired at identical time points within the cardiac cycle, the single-plane and the dual-plane registrations resulted in 3D root-mean-square (rms) errors of 1.60 +/- 0.21 and 0.53 +/- 0.08 mm, respectively. When a 10% timing offset was added between the pre- and the intraoperative acquisitions, the single-plane registration approach resulted in inaccurate registrations in the out-of-plane axis, whereas the dual-plane registration exhibited a 98% success rate with a 3D rms error of 1.33 +/- 0.28 mm. When all potential sources of error were included, namely, the anatomical background, timing offset, and typical errors in the vascular tree reconstruction, the dual-plane registration performed at 94% with an accuracy of 2.19 +/- 0.77 mm.

Fast retrospectively gated quantitative four-dimensional (4D) cardiac micro computed tomography imaging of free-breathing mice

Objective:We sought to demonstrate retrospectively gated dynamic 3D cardiac micro computed tomography (CT) of free-breathing mice. Materials and Methods:Five C57Bl6 mice were scanned using a cone-beam scanner with a slip-ring-mounted flat-panel detector. After the injection of an intravascular iodinated contrast agent, projection images were acquired over the course of 50 seconds, while the scanner rotated through 10 complete rotations. The mouse respiratory and electrocardiogram signals were recorder simultaneously with image acquisition. After acquisition, the projection images were retrospectively sorted into projections belonging to different cardiac time points, occurring only during expiration. Results:Dynamic 3D cardiac images, with isotropic 150-&mgr;m voxel spacing, were reconstructed at 12-millisecond intervals throughout the cardiac cycle in all mice. The average ejection fraction and cardiac output were 58.2 ± 4.6% and 11.4 ± 1.3 mL/min, respectively. The measured entrance dose for the entire scan was 28 cGy. Repeat scans of the same animals showed that intrasubject variability was smaller than intersubject variability. Conclusions:We have developed a high-resolution micro computed tomography method for evaluating the cardiac function and morphology of free-breathing mice in acquisition times shorter than 1 minute.

Prospective respiratory-gated micro-CT of free breathing rodents.

Microcomputed tomography (Micro-CT) has the potential to noninvasively image the structure of organs in rodent models with high spatial resolution and relatively short image acquisition times. However, motion artifacts associated with the normal respiratory motion of the animal may arise when imaging the abdomen or thorax. To reduce these artifacts and the accompanying loss of spatial resolution, we propose a prospective respiratory gating technique for use with anaesthetized, free-breathing rodents. A custom-made bed with an embedded pressure chamber was connected to a pressure transducer. Anaesthetized animals were placed in the prone position on the bed with their abdomens located over the chamber. During inspiration, the motion of the diaphragm caused an increase in the chamber pressure, which was converted into a voltage signal by the transducer. An output voltage was used to trigger image acquisition at any desired time point in the respiratory cycle. Digital radiographic images were acquired of anaesthetized, free-breathing rats with a digital radiographic system to correlate the respiratory wave form with respiration-induced organ motion. The respiratory wave form was monitored and recorded simultaneously with the x-ray radiation pulses, and an imaging window was defined, beginning at end expiration. Phantom experiments were performed to verify that the respiratory gating apparatus was triggering the micro-CT system. Attached to the distensible phantom were 100μm diameter copper wires and the measured full width at half maximum was used to assess differences in image quality between respiratory-gated and ungated imaging protocols. This experiment allowed us to quantify the improvement in the spatial resolution, and the reduction of motion artifacts caused by moving structures, in the images resulting from respiratory-gated image acquisitions. The measured wire diameters were 0.135mm for the stationary phantom image, 0.137mm for the image gated at end deflation, 0.213mm for the image gated at peak inflation, and 0.406mm for the ungated image. Micro-CT images of anaesthetized, free-breathing rats were acquired with a General Electric Healthcare eXplore RS in vivo micro-CT system. Images of the thorax were acquired using the respiratory cycle-based trigger for the respiratory-gated mode. Respiratory gated-images were acquired at inspiration and end expiration, during a period of minimal respiration-induced organ motion. Gated images were acquired with a nominal isotropic voxel spacing of 44μm in 20-25min (80kVp, 113mAs, 300ms imaging window per projection). The equivalent ungated acquisitions were 11min in length. We observed improved definition of the diaphragm boundary and increased conspicuity of small structures within the lungs in the gated images, when compared to the ungated acquisitions. In this work, we have characterized the externally monitored respiratory wave form of free-breathing, anaesthetized rats and correlated the respiration-induced organ motion to the respiratory cycle. We have shown that the respiratory pressure wave form is an excellent surrogate for the radiographic organ motion. This information facilitates the definition of an imaging window at any phase of the breathing cycle. This approach for prospectively gated micro-CT can provide high quality images of anaesthetized free-breathing rodents.

A high‐resolution XRII‐based quantitative volume CT scanner

A laboratory volume CT scanner has been developed, with high spatial resolution in all three dimensions, which can be used for quantitative analysis of excised tissue samples in vitro. The system incorporates an x-ray image intensifier, optically coupled to a time-delay integration (TDI) CCD to obtain low-noise and low-scatter projections of the sample volume. A water bath surrounds the sample to equalize the exposure to the image intensifier, thereby reducing the dynamic range of the input signal. The scanner operates in two modes, producing either a single, transverse image through the sample or a three-dimensional image of the sample volume. Spatial resolution is adjustable over the range of 1.2 to 2.8 mm-1. System response is linear over the range -1000 to 3500 Houndsfield units (HU), with an average precision of +/- 80 HU. The precision of geometric measurements in the transverse plane allows circumference measurements to within +/- 0.1 mm. Finally, applications of this technique of nondestructive analysis in biomedical research are discussed.

Implementation of dual‐ and triple‐energy cone‐beam micro‐CT for postreconstruction material decomposition

Micro-CT has become a powerful tool for small animal research, having the ability to obtain high-resolution in vivo and ex vivo images for analyzing bone mineral content, organ vasculature, and bone microarchitecture extraction. The use of exogenous contrast agents further extends the use of micro-CT techniques, but despite advancements in contrast agents, single-energy micro-CT is still limited in cases where two different materials share similar grey-scale intensity values. This study specifically addresses the development of multiple-energy cone-beam micro-CT, for applications where bone must be separated from blood vessels filled with a Pb-based contrast material (Microfil) in ex vivo studies of rodents and tissue specimens. The authors report the implementation of dual- and triple-energy CT algorithms for material-specific imaging using postreconstruction decomposition of micro-CT data; the algorithms were implemented on a volumetric cone-beam micro-CT scanner (GE Locus Ultra). For the dual-energy approach, extrinsic filtration was applied to the x-ray beam to produce spectra with different proportions of x rays above the K edge of Pb. The optimum x-ray tube energies (140 kVp filtered with 1.45 mm Cu and 96 kVp filtered with 0.3 mm Pb) that maximize the contrast between bone and Microfil were determined through numerical simulation. For the triple-energy decomposition, an additional low-energy spectrum (70 kVp, no added filtration) was used. The accuracy of decomposition was evaluated through simulations and experimental verification of a phantom containing a cortical bone simulating material (SB3), Microfil, and acrylic. Using simulations and phantom experiments, an accuracy greater than 95% was achieved in decompositions of bone and Microfil (for noise levels lower than 11 HU), while soft tissue was separated with accuracy better than 99%. The triple-energy technique demonstrated a slightly higher, but not significantly different, decomposition accuracy than the dual-energy technique for the same achieved noise level in the micro-CT images acquired at the multiple energies. The dual-energy technique was applied to the decomposition of an ex vivo rat specimen perfused with Microfil; successful decomposition of the bone and Microfil was achieved, enabling the visualization and characterization of the vasculature both in areas where the vessels traverse soft tissue and when they are surrounded by bone. In comparison, in single energy micro-CT, vessels surrounded by bone could not be distinguished from the cortical bone, based on grey-scale intensity alone. This work represents the first postreconstruction application of material-specific decomposition that directly takes advantage of the K edge characteristics of a contrast material injected into an animal specimen; the application of the technique resulted in automatic, accurate segmentation of 3D micro-CT images into bone, vessel, and tissue components. The algorithm uses only reconstructed images, rather than projection data, and is calibrated by an operator with signal values in regions identified as being comprised entirely of either cortical bone, contrast-enhanced vessel, or soft tissue; these required calibration values are observed directly within reconstructed CT images acquired at the multiple energies. These features facilitate future implementation on existing research micro-CT systems.

Fibroblast growth factor 9 delivery during angiogenesis produces durable, vasoresponsive microvessels wrapped by smooth muscle cells

The therapeutic potential of angiogenic growth factors has not been realized. This may be because formation of endothelial sprouts is not followed by their muscularization into vasoreactive arteries. Using microarray expression analysis, we discovered that fibroblast growth factor 9 (FGF9) was highly upregulated as human vascular smooth muscle cells (SMCs) assemble into layered cords. FGF9 was not angiogenic when mixed with tissue implants or delivered to the ischemic mouse hind limb, but instead orchestrated wrapping of SMCs around neovessels. SMC wrapping in implants was driven by sonic hedgehog–mediated upregulation of PDGFRβ. Computed tomography microangiography and intravital microscopy revealed that microvessels formed in the presence of FGF9 had enhanced capacity to receive flow and were vasoreactive. Moreover, the vessels persisted beyond 1 year, remodeling into multilayered arteries paired with peripheral nerves. This mature physiological competency was attained by targeting mesenchymal cells rather than endothelial cells, a finding that could inform strategies for therapeutic angiogenesis and tissue engineering.